diff options
| author | Ben Sima <ben@bensima.com> | 2025-12-13 13:28:59 -0500 |
|---|---|---|
| committer | Ben Sima <ben@bensima.com> | 2025-12-13 13:28:59 -0500 |
| commit | a14881ddcdd6ce83250c978d9df825c29e8d93c6 (patch) | |
| tree | 68b355d078c05e37e0ac3267f41fdf8656f22b43 /bs/repl.scm | |
| parent | 4d21f170cd1d1df239d7ad00fbf79427769a140f (diff) | |
telegram: fix audio transcription model and prompt order
- Switch from gemini-2.0-flash-001 to gemini-2.5-flash
- Put audio content before text prompt (model was ignoring audio)
- Strengthen prompt to return only transcription
Diffstat (limited to 'bs/repl.scm')
0 files changed, 0 insertions, 0 deletions